mlsanityData to 14 Oct 2026

ConRTF: Edge-Constrained Boundary Distribution Refinement for Realtime TransFormer Table Structure Recognition

arXiv 2607.00734 ↗First posted 1 Jul 2026cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Table Structure Recognition (TSR) aims to recover the row and column layout of tables from document images, a key step in document understanding pipelines. Accurate TSR depends on precise boundary localization: small errors in row or column boundaries can propagate into incorrect cell assignments and structural inconsistencies. Yet detection-based approaches treat table elements as generic objects, ignoring a fundamental property of table layout: rows and columns play structurally distinct roles and their boundaries carry unequal importance. We propose an Edge-constrained Fine-grained Localization loss (EFL) that formalizes this structural asymmetry by encoding table-specific geometric priors into the training objective: row-like elements are supervised with emphasis on their horizontal boundaries, while column-like elements prioritize vertical boundaries. Implemented within a real-time detector with distribution-based boundary refinement (D-FINE), EFL operates during training only and guides boundary refinement toward structurally meaningful adjustments with no change to the inference pipeline. The proposed approach, ConRTF, is also data-efficient, maintaining robust accuracy with as few as 2k--3k annotated tables. Experiments on PubTables-1M and two private datasets show consistent improvements over the optimized baseline and several real-time detectors including RT-DETRv2 and YOLOv10-11, with gains of up to +1.6 GriTS points at equal inference speed.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 29 of its 33 references that are in the map. First posted on arXiv 1 Jul 2026.

Builds on

29 of 33 references.

References · 29

  1. 2020End-to-End Object Detection with Transformerscited by 7,414
  2. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,422
  3. 2022DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detectioncited by 1,319
  4. 2024YOLOv10: Real-Time End-to-End Object Detectioncited by 515
  5. 2020Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detectioncited by 409
  6. 2025YOLOv12: Attention-Centric Real-Time Object Detectorscited by 305
  7. 2019Image-based table recognition: data, model, and evaluationcited by 237
  8. 2021Rethinking Rotated Object Detection with Gaussian Wasserstein Distance Losscited by 156
  9. 2021Learning High-Precision Bounding Box for Rotated Object Detection via Kullback-Leibler Divergencecited by 130
  10. 2021PubTables-1M: Towards comprehensive table extraction from unstructured documentscited by 126
  11. 2020Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detectioncited by 104
  12. 2022TableFormer: Table Structure Understanding with Transformerscited by 94
17 more
  1. 2024RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformercited by 91
  2. 2024D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinementcited by 74
  3. 2023Optimized Table Tokenization for Table Structure Recognitioncited by 34
  4. 2023ARS-DETR: Aspect Ratio-Sensitive Detection Transformer for Aerial Oriented Object Detectioncited by 32
  5. 2022Robust Table Detection and Structure Recognition from Heterogeneous Document Imagescited by 30
  6. 2022GriTS: Grid table similarity metric for table structure recognitioncited by 28
  7. 2022TSRFormer: Table Structure Recognition with Transformerscited by 27
  8. 2024YOLO advances to its genesis: a decadal and comprehensive review of the You Only Look Once (YOLO) seriescited by 22
  9. 2021Beyond Self-Supervision: A Simple Yet Effective Network Distillation Alternative to Improve Backbonescited by 15
  10. 2022Deep learning for table detection and structure recognition: A surveycited by 15
  11. 2023Aligning benchmark datasets for table structure recognitioncited by 15
  12. 2024UniTabNet: Bridging Vision and Language Models for Enhanced Table Structure Recognitioncited by 11
  13. 2025TFLOP: Table Structure Recognition Framework with Layout Pointer Mechanismcited by 9
  14. 2024LORE++: Logical Location Regression Network for Table Structure Recognition with Pre-trainingcited by 8
  15. 2025TABLET: Table Structure Recognition using Encoder-only Transformerscited by 5
  16. 2025RAPTOR: Refined Approach for Product Table Object Recognitioncited by 3
  17. 2025QUEST: Quality-aware Semi-supervised Table Extraction for Business Documentscited by 2

Built on by

Not cited yet.